Instagram analytics reporting for agencies, post by post and story by story

Reels are the one Instagram format that routinely travels past the existing following, and they are also the format a plain engagement rate describes worst. The Reels tab reports plays — opened views with replays included — beside accounts reached, so the distance between the two is visible instead of buried in one number. Every reel card carries its own length in seconds, the average watch time measured against that length, the watched-through share, the replay count and how much of its reach fell outside the following. A fourteen-second reel with an eleven-second average is a different outcome from a forty-five-second reel with the same average, and only one of the two is a hook problem.

The table beneath the cards ranks every reel by plays and keeps saves and shares in the same row, because a reel that travels has usually earned both. Average watch time is always reported against the reel's own length rather than against a cross-account figure, which is the only honest way to read it: there is no universal number of seconds that makes a reel good, and quoting one to a client invites a comparison the data cannot support. Set the format filter to something other than Reels and the tab says so with a written empty state, rather than showing an empty panel that reads like a loading failure.

A Story is a sequence, and reporting it as one impressions figure throws away the only thing worth knowing: which frame lost the audience. The Stories tab breaks the Story into its frames, distributes the impressions across them so the frames add up to the Story total exactly, and reports each frame's reach, exits, replies, taps forward, taps back and sticker taps in a single table. A drop-off funnel above the table draws the same shape as a row of bars, and the completion figure is derived as the last frame's reach over the first frame's, so the number on screen always has a derivation somebody can point at.

Frame-level detail moves the conversation from Stories did fine to frame three is where they left. A tap back is a compliment — somebody went to re-read a slide — while a run of taps forward on a text-heavy frame usually means that frame asked for too much reading. Replies and sticker taps are counted on their own because they are the two things a Story can earn that a feed post cannot. The tab also states in writing that frame reach counts the accounts that saw that individual frame, so it sits under the Story's own accounts-reached figure, which counts anyone who saw any frame at all.

A like says somebody enjoyed a post. A save says they expect to need it again, and a share says they were willing to attach their own name to it. Those two are the signals that carry content past the accounts already following, so the Engagement tab reports them per format as saves per 1,000 accounts reached and shares per 1,000 accounts reached — normalised, so a format with far more reach cannot win on volume alone. The rates sit beside the raw counts, and the tab names the format currently earning the fewest saves per thousand, because a weak spot that has to be spotted usually is not.

Underneath, engagements break into likes, comments, saves and shares as a stacked bar per format that adds up exactly to the engagement total, with a totals row across every format. That decomposition is what makes an engagement rate legible: two months can post an identical rate while one was carried by likes on a photograph and the other by saves on a how-to carousel, and only the second tells anybody anything about next month. Engagement rate itself is calculated on accounts reached and judged against the organic band Instagram content typically sits in, never against a blended figure borrowed from another network.

The profile is the only place a business account can actually be contacted, which makes the actions taken there the closest thing organic Instagram has to a conversion. The Profile activity tab reports profile visits beside website taps, email taps, call taps, direction taps and follows from the profile, each as a count and as a share of the visits that produced it. Reading contact taps against profile visits rather than against reach is a deliberate choice: somebody has to open the profile before they can tap anything at all, and dividing by reach quietly makes every account look worse than it is.

The action cards make the funnel legible at a glance — how many people opened the profile, how many went to the link, how many made contact, how many followed. Follows from the profile are reported as part of the account's follows for the period rather than as an extra pool stacked on top, so this tab and the Growth tab can be read together without a reconciliation step in between. For a client with a physical location, direction taps and call taps are frequently the two lines that justify the retainer, which is exactly why they are reported instead of folded into an other actions total.

The Growth tab reports follows and unfollows as separate daily series with the net between them, a churn rate measured against the follower level, and a day-by-day ledger listing follows, unfollows, net change and the follower count for every day in the window. Keeping the two flows apart matters because a net figure on its own cannot tell a quiet month from a loud one that shed almost as many followers as it won. The tab also states that follower flows are account-level — Instagram does not attribute a follow to a single post — so the format filter narrows the content on screen and never the audience behind it.

Beside the flows, a follower-activity heat map lays out seven days against twenty-four hours for this account's own followers, with the busiest slot named in words underneath so the chart is not the only way to read it. The Audience tab carries the same shape as the three busiest hours for each weekday, next to age bands split by the account's gender mix, the cities the following is concentrated in, and the country breakdown. Between them they answer the two questions a publishing calendar rests on: who this audience actually is, and when they are likely to be looking at their phone.